PROJECT SUMMARY/ABSTRACT The ability of children to walk effectively is essential to their physical health and general well-being. Unfortunately, many children with cerebral palsy (CP), the most common cause of pediatric physical disability, have difficulty walking and completing balance-intensive weight-bearing tasks. This leads to children with CP engaging in levels of habitual physical activity that are well below guidelines and those of children without disabilities, which in turn contributes to many secondary conditions, including metabolic dysfunction and cardiovascular disease. The pathological gait patterns and reduced stability caused by CP are due, in part, to diminished function of the ankle plantar- and dorsi-flexor muscles. The negative relationship between the energy cost of walking and amount of accumulated physical activity in children with CP underscores the need for new strategies to drastically improve walking economy in this population. To meet this goal, our proposal aims to evaluate the novel use of powered ankle assistance from a wearable exoskeleton to improve walking economy across varied terrain in children with CP. The first specific aim is to quantify how the magnitude of powered plantar-flexor assistance affects the energetics and mechanics of over-ground walking at self-selected speeds, treadmill walking at set speeds, and stair stepping at a constant rate in CP. It is hypothesized that an optimal level of assistance will be identified for each task that maximizes improvements in energy cost of transport, posture, and positive ankle power compared to unassisted walking conditions. The second specific aim is to evaluate how training frequency and patient characteristics affect clinical and biomechanical gait outcomes across repeated over-ground walking sessions with powered ankle assistance in CP. Individuals with gait deficits from CP will be randomly selected to complete 30 minutes of over-ground gait training with individually-tuned plantar-flexor assistance 2x/week for 4 weeks or 4x/week for 2 weeks. Walking economy and speed will be assessed at the beginning, middle, and end of the training period. It is hypothesized that higher frequency training and total powered walking time will be positively associated with gait outcomes, while GMFCS level and spasticity severity will be negatively associated with gait outcomes. The fundamental knowledge gained from this proposal will result in safe and effective control parameters for powered plantar-flexor assistance during over-ground and stair walking in children with CP, and provide insight on training and participant characteristics that can inform the design of future intervention studies. Through supporting student research involvement, this Academic Research Enhancement Award will help grow our new Center for Bioengineering and Bioengineering PhD program at Northern Arizona University.